Description:
Strange crash in pushbuild in 6.0-rpl. The failures happened consistently on win2003-x86/embedded in completion_type_func and innodb_autoinc_lock_mode_func, in every push from https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=bzr_mysql-6.0-rpl&order=61 to https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=bzr_mysql-6.0-rpl&order=66 . It also happened in 5.1-rpl, between the same changesets. It did not happen at a ny other platform, time, or tree.
So it may look like the bug is already fixed. However, there are two candidate pushes that could have fixed it: https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=bzr_mysql-6.0-rpl&order=67 and https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=bzr_mysql-6.0-rpl&order=68 . None of these changed the server (only tests). So the crash can probably still be triggered.
The error message is:
main.completion_type_func [ fail ]
CURRENT_TEST: main.completion_type_func
InnoDB: The first specified data file e:\pb\bzr_mysql-6.0-rpl\66\mysql-6.0.7-alpha-pb66\mysql-test\var-embedded-7\2\mysqld.1\data\ibdata1 did not exist:
InnoDB: a new database to be created!
080910 13:43:53 InnoDB: Setting file e:\pb\bzr_mysql-6.0-rpl\66\mysql-6.0.7-alpha-pb66\mysql-test\var-embedded-7\2\mysqld.1\data\ibdata1 size to 10 MB
InnoDB: Database physically writes the file full: wait...
080910 13:43:53 InnoDB: Log file e:\pb\bzr_mysql-6.0-rpl\66\mysql-6.0.7-alpha-pb66\mysql-test\var-embedded-7\2\mysqld.1\data\ib_logfile0 did not exist: new to be created
InnoDB: Setting log file e:\pb\bzr_mysql-6.0-rpl\66\mysql-6.0.7-alpha-pb66\mysql-test\var-embedded-7\2\mysqld.1\data\ib_logfile0 size to 5 MB
InnoDB: Database physically writes the file full: wait...
080910 13:43:53 InnoDB: Log file e:\pb\bzr_mysql-6.0-rpl\66\mysql-6.0.7-alpha-pb66\mysql-test\var-embedded-7\2\mysqld.1\data\ib_logfile1 did not exist: new to be created
InnoDB: Setting log file e:\pb\bzr_mysql-6.0-rpl\66\mysql-6.0.7-alpha-pb66\mysql-test\var-embedded-7\2\mysqld.1\data\ib_logfile1 size to 5 MB
InnoDB: Database physically writes the file full: wait...
InnoDB: Doublewrite buffer not found: creating new
InnoDB: Doublewrite buffer created
InnoDB: Creating foreign key constraint system tables
InnoDB: Foreign key constraint system tables created
mysqltest got exception 0xc0000005
read_command_buf at 005D7EB0=
conn->name at 066DC090=
conn->cur_query at 00000000=
Attempting backtrace...
101EE60A LIBMYSQLD.dll!innobase_release_temporary_latches()[ha_innodb.cc:549]
101F49ED LIBMYSQLD.dll!ha_innobase::close()[ha_innodb.cc:2486]
100FAF23 LIBMYSQLD.dll!closefrm()[table.cc:2076]
10095BEB LIBMYSQLD.dll!close_cached_tables()[sql_base.cc:902]
100960D6 LIBMYSQLD.dll!table_def_free()[sql_base.cc:277]
10029FAC LIBMYSQLD.dll!clean_up()[mysqld.cc:1336]
1002A8BC LIBMYSQLD.dll!end_embedded_server()[lib_sql.cc:570]
1001402E LIBMYSQLD.dll!mysql_server_end()[libmysql.c:204]
0040F214 mysqltest_embedded.exe!main()[mysqltest.cc:7862]
0043933A mysqltest_embedded.exe!mainCRTStartup()[crt0.c:259]
77E4F38C kernel32.dll!TerminateThread()
- saving 'e:/pb/bzr_mysql-6.0-rpl/66/mysql-6.0.7-alpha-pb66/mysql-test/var-embedded-7/2/log/main.completion_type_func/' to 'e:/pb/bzr_mysql-6.0-rpl/66/mysql-6.0.7-alpha-pb66/mysql-test/var-embedded-7/log/main.completion_type_func/'
Retrying test, attempt(2/3)...
cygpath -m e:/pb/bzr_mysql-6.0-rpl/66/mysql-6.0.7-alpha-pb66/libmysqld/examples/relwithdebinfo/mysqltest_embedded.exe
cygpath -m e:/pb/bzr_mysql-6.0-rpl/66/mysql-6.0.7-alpha-pb66/mysql-test/var-embedded-7/2/log/current_test
How to repeat:
https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=bzr_mysql-6.0-rpl&order=66 win2003-x86/embedded
xref: http://tinyurl.com/4x3p9g
Suggested fix:
The push at https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=bzr_mysql-6.0-rpl&order=67 modified the two failing tests, changing some queries from lowercase to uppercase. That's the only thing that seems remotely related to the crash. I suggest reverting the changset of https://intranet.mysql.com/secure/pushbuild/showpush.pl?dir=bzr_mysql-6.0-rpl&order=67 in these two files and see if the error comes back.
